Improving Diagnostic Processes in the Epilepsy Monitoring Unit

Evaluation in the epilepsy monitoring unit (EMU) is deemed the gold standard for the definitive diagnosis of epilepsy, allowing the formulation of clinico-electrophysiologic correlations to guide pharmacologic and surgical treatments of epilepsy.¹ Optimal management and testing of patients during seizures in the EMU is paramount because it will lead to improved data recording and ultimately reduce any potential risk to the patient. Although long-term video EEG in the EMU is generally safe, vast differences in the processes of care are evident across institutions.²

Consensus statements regarding EMU patient safety as well as a recent feasibility study implementing an ictal testing battery exist.^3,4 In 2016, a taskforce appointed by the International League Against Epilepsy (ILAE) consisting of the Commission on European Affairs and the European Epilepsy Monitoring Unit Association developed a protocol for testing patients during and after seizures in the EMU. A consensus procedure was formulated via a modified Delphi approach based on expert opinion and experience of members from the taskforce.⁴ Despite such a consensus statement, variation in protocols and procedures still exists between units, and studies looking at the uptake of this statement or other standardized procedures are lacking.⁵

In this issue of Neurology: Clinical Practice, O'Kula et al.⁶ developed and implemented a standardized ictal examination in the EMU in a single level 4 epilepsy center. In this study, the authors identified gaps (redundancies, lack of rigorous postictal testing, not written to prioritize safety) in their standardized EMU ictal and postictal assessment and developed a tool (adapted from the 2016 ILAE consensus standards) to evaluate EMU patients. The authors also explored the fidelity of the proposed new testing battery and challenges during implementation. The authors made efforts to improve the process through periodic education, assessment, and improvement.

O'Kula et al. initially used baseline sampling of 20 ictal examinations (10 convulsive and 10 nonconvulsive) that were performed by nurses, EEG technicians, and physicians over a period of 2 months. Errors and omissions were identified by 3 neurologist reviewers. Based on a root cause analysis, failure to understand the rationale for the assessment's components was the main reason to affect optimal testing. Applying local process analysis, the authors developed a standardized, streamlined ictal examination while drawing from the 2016 ILAE consensus guidelines that tailored a validated assessment to their needs. A team of 6 neurologists, 4 neurology house staff, 2 EEG technicians, and 2 nurses developed the intervention. Training and educational sessions were executed through role-playing exercises and distribution of ictal examination cue cards. After 6 months, the first revision was implemented to address the need for seizure-specific assessment and patient-specific considerations. In 1 year, new staff were trained, and dissemination of the assessment tool was accomplished using electronic medical pathways. Evaluations of the intervention by at least 2 physicians in the first month and every 3 months by assessing consecutive convenience samples of 10 convulsive and 10 nonconvulsive seizures at each time point according to a rubric were performed (100 ictal examinations reviewed in total). Despite the endeavors of the authors for process improvement through root cause analysis and education of all stakeholders, the fidelity of the EMU ictal testing was not satisfactory. Small improvements in testing adherence for convulsive seizures were seen postintervention, but the trend was not sustained. Postintervention, there was no sustained trend for ictal or postictal assessment.

The authors performed a rigorous assessment of ictal examination fidelity across various aspects of the examination and seizure types. Although ictal assessment performance did not improve with any significance over time, this study characterized important barriers to ictal testing and showed that a simple educational intervention alone is likely inadequate. Other barriers persisted that were not addressed by the educational interventions such as examiner fatigue, workflow not promoting interdisciplinary collaboration (nurses not typically being a part of EMU rounds), and frequent rotation of house staff. Postictal testing was also inadequate throughout the study in part due to associating convulsive seizures with prolonged duration of unresponsiveness, regardless of rescue medication being administered or not. Postictal testing for nonconvulsive seizures was insufficient. Staff did not complete postictal testing, and rather monitored the EEG, assessed vitals, or substituted the test for a conversation with the patient.

The diagnostic process via the ictal and postictal examination is a refined and interactive process combining patient safety and eliciting information that has clinical significance. Although assimilating pertinent facts and communicating a diagnosis to a patient is often left to the physician, the diagnostic process ideally is a multifaceted partnership, i.e., a team endeavor as identified by the Institute of Medicine (IOM).⁷ The IOM highlights that it is important that performance in diagnostic processes is addressed, improved on, and that competencies are maintained, especially in the following areas: clinical reasoning, teamwork, communication with patients, families, and other health care professionals, appropriate use of testing and application of results in decision making, and use of health information technology.⁸ Training of EMU staff in the service of accurate and effective patient care is important and should be evidence based. It is vital that educational interventions accommodate different learning styles, e.g., include mechanisms of immediate feedback, use competency-based evaluations to assess performance, and place larger emphasis on self-directed learning.⁸ Future improved versions of the ictal assessment should also consider incorporating more individualized patient evaluations.

Author Contributions

C.-S. Kwon and N. Jetté: drafting/revising the manuscript.

Study Funding

No targeted funding reported.

Disclosure

C.-S. Kwon receives funding from the Leon Levy Fellowship. N. Jetté receives grant funding paid to her institution for grants unrelated to this work from the National Institute of Neurological Disorders and Stroke (National Institutes of Health [NIH] U24NS107201 and NIH IU54NS100064), Patient-Centered Outcomes Research Institute, and Alberta Health. She also receives an honorarium for her work as an Associate Editor of Epilepsia. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.